Pub Date : 2024-10-17DOI: 10.1109/OJIES.2024.3483552
Ander González-Docasal;Jon Alonso;Jon Olaizola;Mikel Mendicute;María Patricia Franco;Arantza del Pozo;Daniel Aguinaga;Aitor Álvarez;Eduardo Lleida
This work introduces the design and assessment of a voice-controlled elevator system aimed at facilitating touchless interaction between users and hardware, thereby minimizing contact and improving accessibility for individuals with disabilities. The research distinguishes three distinct deployment scenarios—on cloud, on edge, and embedded—with the ultimate goal of integrating the entire system into a low-resource environment on a custom carrier board. An objective evaluation measured acoustic conditions rigorously using a dataset of 2900 audio files recorded inside a laboratory elevator cabin featuring two internal coatings, five audio input devices, and under four distinct noise conditions. The study evaluated the performance of two Automatic Speech Recognition systems: Google's Speech-to-Text API and a Kaldi model adapted for this task, deployed using Vosk. In addition, latency times for these transcribers and two communication protocols were measured to enhance efficiency. Finally, two subjective evaluations on clean and noisy conditions were conducted simulating a real world scenario. The results, yielding 84.7 and 77.2 points, respectively, in a System Usability Scale questionnaire, affirm the reliability of the presented prototype for industrial deployment.
{"title":"Design and Evaluation of a Voice-Controlled Elevator System to Improve the Safety and Accessibility","authors":"Ander González-Docasal;Jon Alonso;Jon Olaizola;Mikel Mendicute;María Patricia Franco;Arantza del Pozo;Daniel Aguinaga;Aitor Álvarez;Eduardo Lleida","doi":"10.1109/OJIES.2024.3483552","DOIUrl":"https://doi.org/10.1109/OJIES.2024.3483552","url":null,"abstract":"This work introduces the design and assessment of a voice-controlled elevator system aimed at facilitating touchless interaction between users and hardware, thereby minimizing contact and improving accessibility for individuals with disabilities. The research distinguishes three distinct deployment scenarios—on cloud, on edge, and embedded—with the ultimate goal of integrating the entire system into a low-resource environment on a custom carrier board. An objective evaluation measured acoustic conditions rigorously using a dataset of 2900 audio files recorded inside a laboratory elevator cabin featuring two internal coatings, five audio input devices, and under four distinct noise conditions. The study evaluated the performance of two Automatic Speech Recognition systems: Google's Speech-to-Text API and a Kaldi model adapted for this task, deployed using Vosk. In addition, latency times for these transcribers and two communication protocols were measured to enhance efficiency. Finally, two subjective evaluations on clean and noisy conditions were conducted simulating a real world scenario. The results, yielding 84.7 and 77.2 points, respectively, in a System Usability Scale questionnaire, affirm the reliability of the presented prototype for industrial deployment.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"1239-1250"},"PeriodicalIF":5.2,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10721366","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142679290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In robotic assembly of flexible flat cables (FFCs), a unique challenge is the inherent difficulty in manipulating such flexible objects compared to their rigid counterparts and the precise estimation of the cable pose. This work proposes a framework that combines object pose estimation using computer-aided design (CAD) models and multiview fusion to perform precise FFC assembly. Our key insight is that a multiview fusion combined with pretrained 6-D pose estimation models offers a more flexible and precise object pose estimation. In a series of experiments involving FFC insertion tasks requiring assembly tolerances down to 0.1 mm, our approach achieves an insertion success rate of 399 out of 400 total attempts. Furthermore, the assembly tasks include the releasing and securing of FFCs from cable connectors, where the system is successful in 200 out of 200 trials. We have also demonstrated the generalization capability of the methodology by successfully completing insertion tasks for common electronic cables like DisplayPort and USB-A, achieving 199 successes in 200 trials. The results not only validate the feasibility of the proposed approach, but also demonstrate its robustness for real-world industrial applications.
{"title":"On Robust Assembly of Flexible Flat Cables Combining CAD and Image Based Multiview Pose Estimation and a Multimodal Robotic Gripper","authors":"Junbang Liang;Joao Buzzatto;Bryan Busby;Haodan Jiang;Saori Matsunaga;Rintaro Haraguchi;Toshisada Mariyama;Bruce A. MacDonald;Minas Liarokapis","doi":"10.1109/OJIES.2024.3467171","DOIUrl":"https://doi.org/10.1109/OJIES.2024.3467171","url":null,"abstract":"In robotic assembly of flexible flat cables (FFCs), a unique challenge is the inherent difficulty in manipulating such flexible objects compared to their rigid counterparts and the precise estimation of the cable pose. This work proposes a framework that combines object pose estimation using computer-aided design (CAD) models and multiview fusion to perform precise FFC assembly. Our key insight is that a multiview fusion combined with pretrained 6-D pose estimation models offers a more flexible and precise object pose estimation. In a series of experiments involving FFC insertion tasks requiring assembly tolerances down to 0.1 mm, our approach achieves an insertion success rate of 399 out of 400 total attempts. Furthermore, the assembly tasks include the releasing and securing of FFCs from cable connectors, where the system is successful in 200 out of 200 trials. We have also demonstrated the generalization capability of the methodology by successfully completing insertion tasks for common electronic cables like DisplayPort and USB-A, achieving 199 successes in 200 trials. The results not only validate the feasibility of the proposed approach, but also demonstrate its robustness for real-world industrial applications.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"1104-1114"},"PeriodicalIF":5.2,"publicationDate":"2024-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10693648","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142397089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.1109/OJIES.2024.3461949
Praneet Amitabh;Dimitar Bozalakov;Frederik De Belie
In this article, a novel hybrid model of an induction machine is proposed that can emulate the response of a machine with a faulty bearing. The idea behind developing such a topology is to have the response quite close to that from a real asset while keeping it computationally efficient. The aim is to develop an accurate and efficient model, akin to digital twins, which have the potential for real-time operation. Therefore, the model is divided into two parts. One is a physics-based model that takes fundamental equations and motor construction parameters to yield an intermediate response. All the major parameters are taken into account such that the fundamental component comes quite close to that of the real asset and the bearing fault signature comes in the same order. These signatures are quite small and some small parasitic effects or the assumptions taken while simplifying the model might not impact the fundamental component that much but they alter the signature's amplitude quite significantly. One way is to model all the parasitic effects, which might increase the computation effort significantly. Another way is to take all the parasitic effects altogether and bridge the difference using a statistical approach which is developed using experimental data. Therefore, the current measurements were taken for several bearings with different fault severity. These measurements are processed and quantified such that the net outcome can express the evolution of the signature with increasing fault severity. The same is done for the data generated using the physics-based model. Finally, the difference in the responses is reduced using the neural network such that it can mimic real-world machine behavior closely. The analytical model followed by statistical adjustment overall is considered a hybrid model. The ultimate goal of this methodology is to generate extensive datasets encompassing diverse operating conditions that can be used further to estimate the health of the bearing and possibly be used for training predictive algorithms to estimate bearing RUL in motors. The proposed methodology is developed for the machine operating at 1000 and 1500 RPM and is validated for three different operating speeds.
{"title":"Hybrid Modeling of an Induction Machine to Support Bearing Diagnostics","authors":"Praneet Amitabh;Dimitar Bozalakov;Frederik De Belie","doi":"10.1109/OJIES.2024.3461949","DOIUrl":"https://doi.org/10.1109/OJIES.2024.3461949","url":null,"abstract":"In this article, a novel hybrid model of an induction machine is proposed that can emulate the response of a machine with a faulty bearing. The idea behind developing such a topology is to have the response quite close to that from a real asset while keeping it computationally efficient. The aim is to develop an accurate and efficient model, akin to digital twins, which have the potential for real-time operation. Therefore, the model is divided into two parts. One is a physics-based model that takes fundamental equations and motor construction parameters to yield an intermediate response. All the major parameters are taken into account such that the fundamental component comes quite close to that of the real asset and the bearing fault signature comes in the same order. These signatures are quite small and some small parasitic effects or the assumptions taken while simplifying the model might not impact the fundamental component that much but they alter the signature's amplitude quite significantly. One way is to model all the parasitic effects, which might increase the computation effort significantly. Another way is to take all the parasitic effects altogether and bridge the difference using a statistical approach which is developed using experimental data. Therefore, the current measurements were taken for several bearings with different fault severity. These measurements are processed and quantified such that the net outcome can express the evolution of the signature with increasing fault severity. The same is done for the data generated using the physics-based model. Finally, the difference in the responses is reduced using the neural network such that it can mimic real-world machine behavior closely. The analytical model followed by statistical adjustment overall is considered a hybrid model. The ultimate goal of this methodology is to generate extensive datasets encompassing diverse operating conditions that can be used further to estimate the health of the bearing and possibly be used for training predictive algorithms to estimate bearing RUL in motors. The proposed methodology is developed for the machine operating at 1000 and 1500 RPM and is validated for three different operating speeds.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"1140-1157"},"PeriodicalIF":5.2,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10681032","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142452675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Reducing the common mode voltage (CMV) fluctuations is crucial in transformer-less (T-less) converters. The modulation modification-based methods inherently increase the steady-state error of the compared currents due to the reduced number of voltage vectors. This error can significantly raise the total harmonic distortion (THD) output current of the inverter. This research presents a strategy of odd virtual vectors based on model-free predictive control using the extended state observer (ESO) to fix the CMV fluctuations and a significant decrease in the THD of the output current. This means the number of CMV stabilizing vectors increases with the linear combination of odd voltage vectors. The proposed method has two advantages over CMV fluctuation reduction schemes that are modulation modification-based: simultaneous control of CMV stabilization and THD reduction in T-less converters, and independence of the controller from system variables and parameters, making it a robust predictive control method. The practical results show that the proposed method, in addition to the complete CMV stabilization and the reduction of the current THD, is completely robust to the changes in the parameters of the ultralocal model and ESO compared to the model-based solutions.
{"title":"Model-Free Predictive Current Controller for Common Mode Voltage Stabilization by Finite odd Virtual Vector set","authors":"Majid Akbari;S. Alireza Davari;Reza Ghandehari;Freddy Flores-Bahamonde;Jose Rodriguez","doi":"10.1109/OJIES.2024.3457835","DOIUrl":"10.1109/OJIES.2024.3457835","url":null,"abstract":"Reducing the common mode voltage (CMV) fluctuations is crucial in transformer-less (T-less) converters. The modulation modification-based methods inherently increase the steady-state error of the compared currents due to the reduced number of voltage vectors. This error can significantly raise the total harmonic distortion (THD) output current of the inverter. This research presents a strategy of odd virtual vectors based on model-free predictive control using the extended state observer (ESO) to fix the CMV fluctuations and a significant decrease in the THD of the output current. This means the number of CMV stabilizing vectors increases with the linear combination of odd voltage vectors. The proposed method has two advantages over CMV fluctuation reduction schemes that are modulation modification-based: simultaneous control of CMV stabilization and THD reduction in T-less converters, and independence of the controller from system variables and parameters, making it a robust predictive control method. The practical results show that the proposed method, in addition to the complete CMV stabilization and the reduction of the current THD, is completely robust to the changes in the parameters of the ultralocal model and ESO compared to the model-based solutions.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"1042-1057"},"PeriodicalIF":5.2,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10675355","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142175875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-06DOI: 10.1109/OJIES.2024.3455264
Maryam Assafo;Peter Langendoerfer
Tool condition monitoring (TCM) is crucial to ensure good quality products and avoid downtime. Machine learning has proven to be vital for TCM. However, existing works are predominately based on supervised learning, which hinders their applicability in real-world manufacturing settings, where data labeling is cumbersome and costly with in-service machines. Additionally, the existing unsupervised solutions mostly handle binary decision-based TCM which is unable to fully reflect the dynamics of tool wear progression. To address these issues, we propose different unsupervised and semisupervised five-class tool wear recognition frameworks to handle fully unlabeled and partially labeled data, respectively. The underlying methods include Laplacian score, sparse autoencoder (SAE), stacked SAE (SSAE), self-organizing map, Softmax, support vector machine, and random forest. For the semisupervised frameworks, we considered designs where labeled data influence only feature learning, classifier building, or both. We also investigated different training configurations of SSAE regarding the supervision level. We applied the frameworks on two run-to-failure datasets of milling tools, recorded using a microphone and an accelerometer. Single sensor and multisensor data under different percentages of labeled training data were considered in the evaluation. The results showed which of the frameworks led to the best predictive performance under which data settings, and highlighted the significance of sensor fusion and discriminative feature representations in combating the unavailability and scarcity of labels, among other findings. The highest macro-F1 achieved for the two datasets with fully unlabeled data reached 87.52% and 75.80%, respectively, and over 90% when only 25% of the training observations were labeled.
{"title":"Unsupervised and Semisupervised Machine Learning Frameworks for Multiclass Tool Wear Recognition","authors":"Maryam Assafo;Peter Langendoerfer","doi":"10.1109/OJIES.2024.3455264","DOIUrl":"10.1109/OJIES.2024.3455264","url":null,"abstract":"Tool condition monitoring (TCM) is crucial to ensure good quality products and avoid downtime. Machine learning has proven to be vital for TCM. However, existing works are predominately based on supervised learning, which hinders their applicability in real-world manufacturing settings, where data labeling is cumbersome and costly with in-service machines. Additionally, the existing unsupervised solutions mostly handle binary decision-based TCM which is unable to fully reflect the dynamics of tool wear progression. To address these issues, we propose different unsupervised and semisupervised five-class tool wear recognition frameworks to handle fully unlabeled and partially labeled data, respectively. The underlying methods include Laplacian score, sparse autoencoder (SAE), stacked SAE (SSAE), self-organizing map, Softmax, support vector machine, and random forest. For the semisupervised frameworks, we considered designs where labeled data influence only feature learning, classifier building, or both. We also investigated different training configurations of SSAE regarding the supervision level. We applied the frameworks on two run-to-failure datasets of milling tools, recorded using a microphone and an accelerometer. Single sensor and multisensor data under different percentages of labeled training data were considered in the evaluation. The results showed which of the frameworks led to the best predictive performance under which data settings, and highlighted the significance of sensor fusion and discriminative feature representations in combating the unavailability and scarcity of labels, among other findings. The highest macro-F1 achieved for the two datasets with fully unlabeled data reached 87.52% and 75.80%, respectively, and over 90% when only 25% of the training observations were labeled.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"993-1010"},"PeriodicalIF":5.2,"publicationDate":"2024-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10668405","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142175872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
While it is a widespread understanding that the sustainability of the global economy requires a transition to a circular economy paradigm where a growing share of the raw materials resources used for the manufacturing of the products are recycled when products reach their end-of-life, still this much-needed transition faces organizational and technical challenges. The key technical and economic bottlenecks are in the automation of disassembly. In this article, we propose a viable functional framework for the systematic analysis, design, and implementation of disassembly cells. This framework consists of two main parts: a systematic categorization of disassembly tasks and a modular and flexible hardware (HW)/software (SW) architecture of a disassembly cell able to implement the disassembly tasks. We analyze and categorize human manipulation when disassembling a common object of daily working activities as a new companion concept to the more common concept of daily life activities. We tested and validated our methodology on the disassembly of a car suspension.
{"title":"A Functional and Practical Taxonomy for the Industrial Implementation of Highly Automated Reverse Manufacturing Cells","authors":"Annagiulia Morachioli;Vladimir Sivtsov;Nicolas Rojas;Fabio Bonsignorio","doi":"10.1109/OJIES.2024.3453900","DOIUrl":"10.1109/OJIES.2024.3453900","url":null,"abstract":"While it is a widespread understanding that the sustainability of the global economy requires a transition to a circular economy paradigm where a growing share of the raw materials resources used for the manufacturing of the products are recycled when products reach their end-of-life, still this much-needed transition faces organizational and technical challenges. The key technical and economic bottlenecks are in the automation of disassembly. In this article, we propose a viable functional framework for the systematic analysis, design, and implementation of disassembly cells. This framework consists of two main parts: a systematic categorization of disassembly tasks and a modular and flexible hardware (HW)/software (SW) architecture of a disassembly cell able to implement the disassembly tasks. We analyze and categorize human manipulation when disassembling a common object of daily working activities as a new companion concept to the more common concept of daily life activities. We tested and validated our methodology on the disassembly of a car suspension.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"1115-1139"},"PeriodicalIF":5.2,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10666886","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142175874","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-05DOI: 10.1109/OJIES.2024.3455239
Thales Augusto Fagundes;Guilherme Henrique Favaro Fuzato;Lucas Jonys Ribeiro Silva;Augusto Matheus dos Santos Alonso;Juan C. Vasquez;Josep M. Guerrero;Ricardo Quadros Machado
Microgrids (MGs) often integrate various energy sources to enhance system reliability, including intermittent methods, such as solar panels and wind turbines. Consequently, this integration contributes to a more resilient power distribution system. In addition, battery energy storage system (BESS) units are connected to MGs to offer grid-supporting services, such as peak shaving, load compensation, power factor quality, and operation during source failures. In this context, an energy management system (EMS) is necessary to incorporate BESS in MGs. Consequently, state-of-charge (SoC) equalization is a common approach to address EMS requirements and balance the internal load among BESS units in MG operation. In this article, we present a comprehensive review of EMS strategies for balancing SoC among BESS units, including centralized and decentralized control, multiagent systems, and other concepts, such as designing nonlinear strategies, optimal algorithms, and categorizing agents into clusters. Moreover, in this article, we discuss alternatives to improve EMS and strategies regarding the topology of power converters, including redundancy-based topology, modular multilevel converter, cascaded-based converter, and hybrid-type systems. In addition, this article explores optimization processes aimed at reducing operational costs while considering SoC equalization. Finally, second-life BESS units are explored as an emerging topic, focusing on their operation within specific power converters topologies to achieve SoC balance.
{"title":"Battery Energy Storage Systems in Microgrids: A Review of SoC Balancing and Perspectives","authors":"Thales Augusto Fagundes;Guilherme Henrique Favaro Fuzato;Lucas Jonys Ribeiro Silva;Augusto Matheus dos Santos Alonso;Juan C. Vasquez;Josep M. Guerrero;Ricardo Quadros Machado","doi":"10.1109/OJIES.2024.3455239","DOIUrl":"10.1109/OJIES.2024.3455239","url":null,"abstract":"Microgrids (MGs) often integrate various energy sources to enhance system reliability, including intermittent methods, such as solar panels and wind turbines. Consequently, this integration contributes to a more resilient power distribution system. In addition, battery energy storage system (BESS) units are connected to MGs to offer grid-supporting services, such as peak shaving, load compensation, power factor quality, and operation during source failures. In this context, an energy management system (EMS) is necessary to incorporate BESS in MGs. Consequently, state-of-charge (SoC) equalization is a common approach to address EMS requirements and balance the internal load among BESS units in MG operation. In this article, we present a comprehensive review of EMS strategies for balancing SoC among BESS units, including centralized and decentralized control, multiagent systems, and other concepts, such as designing nonlinear strategies, optimal algorithms, and categorizing agents into clusters. Moreover, in this article, we discuss alternatives to improve EMS and strategies regarding the topology of power converters, including redundancy-based topology, modular multilevel converter, cascaded-based converter, and hybrid-type systems. In addition, this article explores optimization processes aimed at reducing operational costs while considering SoC equalization. Finally, second-life BESS units are explored as an emerging topic, focusing on their operation within specific power converters topologies to achieve SoC balance.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"961-992"},"PeriodicalIF":5.2,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10666276","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142175873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-03DOI: 10.1109/OJIES.2024.3454010
Tianshi Cheng;Tong Duan;Venkata Dinavahi
Low Earth orbit (LEO) satellite networks, such as SpaceX's Starlink, offer enhanced communication potential for contemporary power grid measurement and control. Yet, the dynamic nature of these networks complicates their modeling and simulation. This study introduces a modular, data-oriented digital twin framework for real-time simulation of wide-area ac–dc grids with LEO satellite networks. The framework integrates RustSat for satellite tracking, SatSDN with MiniNet for SDN simulations, and entity-component-system (ECS)-Grid for real-time power system simulation. It features a data-centric design using an ECS framework with a structure-of-arrays memory layout, optimizing cache efficiency and computational performance, and offers high extensibility for interdisciplinary simulations. This marks the initial effort to develop a digital twin for real-time co-simulation of large-scale power systems and LEO satellite constellation networks. Evaluations on a wide-area synthetic ac–dc system with multiple satellite network types confirm the efficiency and precision of our approach, underscoring its potential in bridging LEO satellite networks with power system applications.
{"title":"Real-Time Cyber-Physical Digital Twin for Low Earth Orbit Satellite Constellation Network Enhanced Wide-Area Power Grid","authors":"Tianshi Cheng;Tong Duan;Venkata Dinavahi","doi":"10.1109/OJIES.2024.3454010","DOIUrl":"10.1109/OJIES.2024.3454010","url":null,"abstract":"Low Earth orbit (LEO) satellite networks, such as SpaceX's Starlink, offer enhanced communication potential for contemporary power grid measurement and control. Yet, the dynamic nature of these networks complicates their modeling and simulation. This study introduces a modular, data-oriented digital twin framework for real-time simulation of wide-area ac–dc grids with LEO satellite networks. The framework integrates RustSat for satellite tracking, SatSDN with MiniNet for SDN simulations, and entity-component-system (ECS)-Grid for real-time power system simulation. It features a data-centric design using an ECS framework with a structure-of-arrays memory layout, optimizing cache efficiency and computational performance, and offers high extensibility for interdisciplinary simulations. This marks the initial effort to develop a digital twin for real-time co-simulation of large-scale power systems and LEO satellite constellation networks. Evaluations on a wide-area synthetic ac–dc system with multiple satellite network types confirm the efficiency and precision of our approach, underscoring its potential in bridging LEO satellite networks with power system applications.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"1029-1041"},"PeriodicalIF":5.2,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10663871","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142175876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-29DOI: 10.1109/OJIES.2024.3451959
Riccardo Berta;Ali Dabbous;Luca Lazzaroni;Danilo Pietro Pau;Francesco Bellotti
Tiny machine learning technologies are bringing intelligence ever closer to the sensor, thus enabling the key benefits of edge computing (e.g., reduced latency, improved data security, higher energy efficiency, and lower bandwidth consumption, also without the need for constant connectivity). This promises to significantly enhance industrial applications but requires suited development tools to deal with the complexity of the edge technologies and context. We propose an agile Jupyter Python notebook as a simple, manageable tool to efficiently and effectively develop microcontroller-based intelligent imaging classification sensors. The notebook implements a methodology involving hyperparameter tuning and comparison of different shallow and deep learning models, with quantization. It exports TensorFlow Lite models, deployable on several microcontroller families, and optionally exploits the STM32Cube.AI developer cloud service, which allows benchmarking the developed models on a set of real-world tiny hardware target platforms. Assessment concerns various types of metrics, both for machine learning (e.g., accuracy) and embedded systems (e.g., memory footprint, latency, and energy consumption). We have verified the support for development effectiveness and efficiency on four ultralow resolution image-classification datasets, with different levels of input and task complexity. In all cases, the tool was able to build microcontroller-deployment ready, beyond the state-of-the-art models, within 1 h on Google Colab CPUs.
微小的机器学习技术使智能越来越接近传感器，从而实现了边缘计算的主要优势（例如，减少延迟、提高数据安全性、提高能效和降低带宽消耗，而且无需持续连接）。这有望大幅提升工业应用，但需要合适的开发工具来应对边缘技术和环境的复杂性。我们提出了一种灵活的 Jupyter Python 笔记本，作为一种简单、易于管理的工具，用于高效开发基于微控制器的智能成像分类传感器。该笔记本实现的方法涉及超参数调整、不同浅层和深度学习模型的比较以及量化。它输出 TensorFlow Lite 模型，可部署在多个微控制器系列上，并可选择利用 STM32Cube.AI 开发人员云服务，该服务允许在一组真实世界的微型硬件目标平台上对所开发的模型进行基准测试。评估涉及机器学习（如准确性）和嵌入式系统（如内存占用、延迟和能耗）的各类指标。我们在四个超低分辨率图像分类数据集上验证了该支持工具的开发效果和效率，这些数据集具有不同程度的输入和任务复杂性。在所有情况下，该工具都能在谷歌 Colab CPU 上在 1 小时内构建出微控制器部署就绪的、超越最先进模型的模型。
{"title":"Developing a TinyML Image Classifier in an Hour","authors":"Riccardo Berta;Ali Dabbous;Luca Lazzaroni;Danilo Pietro Pau;Francesco Bellotti","doi":"10.1109/OJIES.2024.3451959","DOIUrl":"10.1109/OJIES.2024.3451959","url":null,"abstract":"Tiny machine learning technologies are bringing intelligence ever closer to the sensor, thus enabling the key benefits of edge computing (e.g., reduced latency, improved data security, higher energy efficiency, and lower bandwidth consumption, also without the need for constant connectivity). This promises to significantly enhance industrial applications but requires suited development tools to deal with the complexity of the edge technologies and context. We propose an agile Jupyter Python notebook as a simple, manageable tool to efficiently and effectively develop microcontroller-based intelligent imaging classification sensors. The notebook implements a methodology involving hyperparameter tuning and comparison of different shallow and deep learning models, with quantization. It exports TensorFlow Lite models, deployable on several microcontroller families, and optionally exploits the STM32Cube.AI developer cloud service, which allows benchmarking the developed models on a set of real-world tiny hardware target platforms. Assessment concerns various types of metrics, both for machine learning (e.g., accuracy) and embedded systems (e.g., memory footprint, latency, and energy consumption). We have verified the support for development effectiveness and efficiency on four ultralow resolution image-classification datasets, with different levels of input and task complexity. In all cases, the tool was able to build microcontroller-deployment ready, beyond the state-of-the-art models, within 1 h on Google Colab CPUs.","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"946-960"},"PeriodicalIF":5.2,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10659231","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142175877","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This article addresses the mitigation of dynamic voltage imbalance in series-connected 10 kV silicon carbide (SiC) JBS diodes within a three-level NPC (3L-NPC) converter using active turn- <sc>off</small> delay control across complementary series-connected 10 kV SiC <sc>mosfet</small> s. The implementation of active turn- <sc>off</small> delay control in SiC <sc>mosfet</small> s eliminates the need for passive <inline-formula><tex-math>$RC$</tex-math></inline-formula> snubbers, which otherwise increase the switching <inline-formula><tex-math>$dv/dt$</tex-math></inline-formula> mismatch and snubber current across the diodes. In addition, parasitic base-plate capacitance across <sc>mosfet</small> s and diodes, along with parasitic bus bar and snubber inductance in the commutation path, contribute to turn- <sc>off</small> voltage mismatch and snubber loss in series-connected 10 kV SiC JBS diodes. The mismatch in nonlinear capacitance of series-connected devices ( <sc>mosfet</small> s and diodes) and the nonlinear <sc>mosfet</small> <inline-formula><tex-math>$i$</tex-math></inline-formula> <inline-formula><tex-math>$-$</tex-math></inline-formula> <inline-formula><tex-math>$v_{gs}$</tex-math></inline-formula> curve affect the turn- <sc>on</small> and turn- <sc>off</small> voltage transitions between complementary switching <sc>mosfet</small> s and diodes, leading to variations in turn- <sc>off</small> voltage mismatch and snubber losses. The 3L-NPC converter has eight types of switching transition, complicating the analysis of <inline-formula><tex-math>$RC$</tex-math></inline-formula> snubber design. This complexity is further increased by nonlinear device parameters, parasitic capacitance, and inductance in the commutation path for each of the eight 10 kV SiC <sc>mosfet</small> s and four 10 kV SiC JBS diodes. To address these challenges, this research develops a mathematical model for the switching transition between 10 kV SiC <sc>mosfet</small> s and complementary 10 kV SiC JBS diodes in a two-level clamped inductive switching (CIS) test setup. The model considers the effects of parasitic base-plate capacitance and the absence of an <inline-formula><tex-math>$RC$</tex-math></inline-formula> snubber due to active turn- <sc>off</small> delay control across series-connected SiC <sc>mosfet</small> s. Subsequently, the mathematical model is refined using an iterative algorithm to account for mismatches in nonlinear device capacitance of <sc>mosfet</small> s and diodes, as well as the nonlinear <inline-formula><tex-math>$i$</tex-math></inline-formula> <inline-formula><tex-math>$-$</tex-math></inline-formula> <inline-formula><tex-math>$v_{gs}$</tex-math></inline-formula> curve of <sc>mosfet</small> s during the switching transition of the diode. This refined model is then used to design the <inline-formula><tex-math>$RC$</tex-math></inline-formula> snubber for series-connected 10 kV SiC JBS diodes and to optimize the turn-
本文探讨了在三电平 NPC（3L-NPC）转换器中，利用互补串联的 10 kV SiC mosfets 上的有源关断延迟控制来缓解串联连接的 10 kV 碳化硅（SiC）JBS 二极管的动态电压不平衡问题。在碳化硅镶嵌晶体管中实施有源关断延迟控制，无需使用无源 $RC$ 缓冲器，否则会增加二极管上的开关 $dv/dt$ 失配和缓冲电流。此外，mosfets 和二极管上的寄生基板电容，以及换向路径中的寄生母线和缓冲器电感，都会导致串联 10 kV SiC JBS 二极管的关断电压失配和缓冲器损耗。串联器件（mosfet 和二极管）非线性电容的不匹配以及非线性 mosfet $i$$-$v_{gs}$ 曲线会影响互补开关 mosfet 和二极管之间的导通和关断电压转换，从而导致关断电压不匹配和缓冲器损耗的变化。3L-NPC 转换器有八种类型的开关转换，这使 $RC$ 缓冲器的设计分析变得更加复杂。8 个 10 kV SiC mosfets 和 4 个 10 kV SiC JBS 二极管换向路径中的非线性器件参数、寄生电容和电感进一步增加了这种复杂性。为了应对这些挑战，本研究针对两级箝位电感开关 (CIS) 测试装置中 10 kV SiC mosfets 和互补 10 kV SiC JBS 二极管之间的开关转换建立了一个数学模型。该模型考虑了寄生基板电容的影响，以及由于串联 SiC mosfets 的主动关断延迟控制而不存在 $RC$ 缓冲器的情况。随后，利用迭代算法对数学模型进行了改进，以考虑到晶体管和二极管非线性器件电容的不匹配，以及晶体管在二极管开关转换期间的非线性 $i$$-$v_{gs}$ 曲线。然后，利用这一完善的模型设计串联 10 kV SiC JBS 二极管的 $RC$ 缓冲器，并在两级 CIS 测试台（TB1 和 TB2）上优化互补 10 kV SiC mosfets 的导通栅极电阻。随后，利用 3L-NPC 测试台 3 至 5 的实验结果对设计参数进行了系统调整。本文提供了在各种测试台中设计和分析 $RC$ 缓冲器的简化步骤，并通过实验数据进行了验证。采用最终的 $RC$ 缓冲器设计的 3L-NPC 转换器实现了 99.2% 的效率和 35 V 的关断电压失配。理论模型与实验数据之间的最大误差为 4.8%。
{"title":"Dynamic Voltage Balancing Across Series-Connected 10 kV SiC JBS Diodes in Medium Voltage 3L-NPC Power Converter Having Snubberless Series-Connected 10 kV SiC MOSFETs","authors":"Sanket Parashar;Nithin Kolli;Raj Kumar Kokkonda;Ajit Kanale;Subhashish Bhattacharya;Bantval Jayant Baliga","doi":"10.1109/OJIES.2024.3450509","DOIUrl":"10.1109/OJIES.2024.3450509","url":null,"abstract":"This article addresses the mitigation of dynamic voltage imbalance in series-connected 10 kV silicon carbide (SiC) JBS diodes within a three-level NPC (3L-NPC) converter using active turn-\u0000<sc>off</small>\u0000 delay control across complementary series-connected 10 kV SiC \u0000<sc>mosfet</small>\u0000s. The implementation of active turn-\u0000<sc>off</small>\u0000 delay control in SiC \u0000<sc>mosfet</small>\u0000s eliminates the need for passive \u0000<inline-formula><tex-math>$RC$</tex-math></inline-formula>\u0000 snubbers, which otherwise increase the switching \u0000<inline-formula><tex-math>$dv/dt$</tex-math></inline-formula>\u0000 mismatch and snubber current across the diodes. In addition, parasitic base-plate capacitance across \u0000<sc>mosfet</small>\u0000s and diodes, along with parasitic bus bar and snubber inductance in the commutation path, contribute to turn-\u0000<sc>off</small>\u0000 voltage mismatch and snubber loss in series-connected 10 kV SiC JBS diodes. The mismatch in nonlinear capacitance of series-connected devices (\u0000<sc>mosfet</small>\u0000s and diodes) and the nonlinear \u0000<sc>mosfet</small>\u0000 \u0000<inline-formula><tex-math>$i$</tex-math></inline-formula>\u0000<inline-formula><tex-math>$-$</tex-math></inline-formula>\u0000<inline-formula><tex-math>$v_{gs}$</tex-math></inline-formula>\u0000 curve affect the turn-\u0000<sc>on</small>\u0000 and turn-\u0000<sc>off</small>\u0000 voltage transitions between complementary switching \u0000<sc>mosfet</small>\u0000s and diodes, leading to variations in turn-\u0000<sc>off</small>\u0000 voltage mismatch and snubber losses. The 3L-NPC converter has eight types of switching transition, complicating the analysis of \u0000<inline-formula><tex-math>$RC$</tex-math></inline-formula>\u0000 snubber design. This complexity is further increased by nonlinear device parameters, parasitic capacitance, and inductance in the commutation path for each of the eight 10 kV SiC \u0000<sc>mosfet</small>\u0000s and four 10 kV SiC JBS diodes. To address these challenges, this research develops a mathematical model for the switching transition between 10 kV SiC \u0000<sc>mosfet</small>\u0000s and complementary 10 kV SiC JBS diodes in a two-level clamped inductive switching (CIS) test setup. The model considers the effects of parasitic base-plate capacitance and the absence of an \u0000<inline-formula><tex-math>$RC$</tex-math></inline-formula>\u0000 snubber due to active turn-\u0000<sc>off</small>\u0000 delay control across series-connected SiC \u0000<sc>mosfet</small>\u0000s. Subsequently, the mathematical model is refined using an iterative algorithm to account for mismatches in nonlinear device capacitance of \u0000<sc>mosfet</small>\u0000s and diodes, as well as the nonlinear \u0000<inline-formula><tex-math>$i$</tex-math></inline-formula>\u0000<inline-formula><tex-math>$-$</tex-math></inline-formula>\u0000<inline-formula><tex-math>$v_{gs}$</tex-math></inline-formula>\u0000 curve of \u0000<sc>mosfet</small>\u0000s during the switching transition of the diode. This refined model is then used to design the \u0000<inline-formula><tex-math>$RC$</tex-math></inline-formula>\u0000 snubber for series-connected 10 kV SiC JBS diodes and to optimize the turn-\u0000","PeriodicalId":52675,"journal":{"name":"IEEE Open Journal of the Industrial Electronics Society","volume":"5 ","pages":"1058-1084"},"PeriodicalIF":5.2,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10652239","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142175878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：